Should fix a comment referring to stats collector?
Hi,
Statistics collector has been removed since 5891c7a8ed8f2d3d5, but there
was a comment referring 'statistics collector' in pg_statistic.h.
Note that since the arrays are variable-size, K may be chosen by the
statistics collector.
Should it be modified to 'cumulative statistics system' like manual on
monitoring stats[1]https://www.postgresql.org/docs/current/monitoring-stats.html?
Its title has changed from 'statistics collector' to 'cumulative
statistics system'.
[1]: https://www.postgresql.org/docs/current/monitoring-stats.html
--
Regards,
--
Atsushi Torikoshi
NTT DATA CORPORATION
Attachments:
v1-0001-Modified-a-comment-referring-to-stats-collector.patchtext/x-diff; name=v1-0001-Modified-a-comment-referring-to-stats-collector.patchDownload
From 5be784278e8e7aeeeadf60a772afccda7b59e6e4 Mon Sep 17 00:00:00 2001
From: Atsushi Torikoshi <torikoshia@oss.nttdata.com>
Date: Fri, 29 Jul 2022 21:34:15 +0900
Subject: [PATCH v1] Modified a comment referring to stats collector.
As statistics collector has been removed since 5891c7a8ed8f2d3d5,
comments refering to statistics collector should be modified to
cumulative statistics system.
---
src/include/catalog/pg_statistic.h | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/src/include/catalog/pg_statistic.h b/src/include/catalog/pg_statistic.h
index cdf7448139..d07fd2f93f 100644
--- a/src/include/catalog/pg_statistic.h
+++ b/src/include/catalog/pg_statistic.h
@@ -178,9 +178,9 @@ DECLARE_FOREIGN_KEY((starelid, staattnum), pg_attribute, (attrelid, attnum));
* the K most common non-null values appearing in the column, and stanumbers
* contains their frequencies (fractions of total row count). The values
* shall be ordered in decreasing frequency. Note that since the arrays are
- * variable-size, K may be chosen by the statistics collector. Values should
- * not appear in MCV unless they have been observed to occur more than once;
- * a unique column will have no MCV slot.
+ * variable-size, K may be chosen by the cumulative statistics system.
+ * Values should not appear in MCV unless they have been observed to occur
+ * more than once; a unique column will have no MCV slot.
*/
#define STATISTIC_KIND_MCV 1
base-commit: 59be1c942a47f6c8a4c47d242200fbbf4be59b88
--
2.27.0
On 2022-Jul-29, torikoshia wrote:
Statistics collector has been removed since 5891c7a8ed8f2d3d5, but there was
a comment referring 'statistics collector' in pg_statistic.h.Note that since the arrays are variable-size, K may be chosen by the
statistics collector.Should it be modified to 'cumulative statistics system' like manual on
monitoring stats[1]?
I don't think this refers to the statistics collector process; I
understand it to refer to ANALYZE that captures the data being stored.
Maybe it should just say "K may be chosen at ANALYZE time".
--
Álvaro Herrera PostgreSQL Developer — https://www.EnterpriseDB.com/
"Nadie está tan esclavizado como el que se cree libre no siéndolo" (Goethe)
On 2022-07-30 02:53, Alvaro Herrera wrote:
I don't think this refers to the statistics collector process; I
understand it to refer to ANALYZE that captures the data being stored.
Thanks for the explanation!
Maybe it should just say "K may be chosen at ANALYZE time".
It seems clearer than current one.
--
Regards,
--
Atsushi Torikoshi
NTT DATA CORPORATION
On Mon, Aug 1, 2022 at 09:05:45PM +0900, torikoshia wrote:
On 2022-07-30 02:53, Alvaro Herrera wrote:
I don't think this refers to the statistics collector process; I
understand it to refer to ANALYZE that captures the data being stored.Thanks for the explanation!
Maybe it should just say "K may be chosen at ANALYZE time".
It seems clearer than current one.
Change made in master.
--
Bruce Momjian <bruce@momjian.us> https://momjian.us
EDB https://enterprisedb.com
Only you can decide what is important to you.
On Tue, Oct 31, 2023 at 11:02 PM Bruce Momjian <bruce@momjian.us> wrote:
On Mon, Aug 1, 2022 at 09:05:45PM +0900, torikoshia wrote:
On 2022-07-30 02:53, Alvaro Herrera wrote:
I don't think this refers to the statistics collector process; I
understand it to refer to ANALYZE that captures the data being stored.Thanks for the explanation!
Maybe it should just say "K may be chosen at ANALYZE time".
It seems clearer than current one.
Change made in master.
I believe some redundant wording has been committed.
K may be chosen may be chosen at ANALYZE time.
Attached patch fixes it and with some line adjustments.
--
Bruce Momjian <bruce@momjian.us> https://momjian.us
EDB https://enterprisedb.comOnly you can decide what is important to you.
--
Regards
Junwang Zhao
Attachments:
v1-0001-remove-redundant-wording-in-pg_statistic.h.patchapplication/octet-stream; name=v1-0001-remove-redundant-wording-in-pg_statistic.h.patchDownload
From 4f2c1f3c1f137af391156194d179c80f18c0f97b Mon Sep 17 00:00:00 2001
From: Junwang Zhao <zhjwpku@gmail.com>
Date: Sat, 28 Dec 2024 07:18:56 +0000
Subject: [PATCH v1] remove redundant wording in pg_statistic.h
---
src/include/catalog/pg_statistic.h | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/src/include/catalog/pg_statistic.h b/src/include/catalog/pg_statistic.h
index 041e626390..667782384e 100644
--- a/src/include/catalog/pg_statistic.h
+++ b/src/include/catalog/pg_statistic.h
@@ -183,9 +183,9 @@ DECLARE_FOREIGN_KEY((starelid, staattnum), pg_attribute, (attrelid, attnum));
* the K most common non-null values appearing in the column, and stanumbers
* contains their frequencies (fractions of total row count). The values
* shall be ordered in decreasing frequency. Note that since the arrays are
- * variable-size, K may be chosen may be chosen at ANALYZE time. Values should
- * not appear in MCV unless they have been observed to occur more than once;
- * a unique column will have no MCV slot.
+ * variable-size, K may be chosen at ANALYZE time. Values should not appear
+ * in MCV unless they have been observed to occur more than once; a unique
+ * column will have no MCV slot.
*/
#define STATISTIC_KIND_MCV 1
--
2.39.5
On Sat, Dec 28, 2024 at 03:25:46PM +0800, Junwang Zhao wrote:
I believe some redundant wording has been committed.
K may be chosen may be chosen at ANALYZE time.
Attached patch fixes it and with some line adjustments.
Indeed. Fixed, thanks!
--
Michael